Hacker News Reader

Comment by mikeknoop | Hacker News Reader

The Panic status board is near to my heart. Back in 2010, their launch blog post was something that inspired me to try and build products for people instead of freelancing for a living.

It's really interesting/sad to read their second reason for shutting down:

> pro users are more likely to want a larger number of integrations with new services and data sources, something that’s hard to provide with limited revenue, which left the app “close but not quite” for many users

Because this is exactly what Zapier, the company I co-founded a year later, provides for free to other companies/products. Integrate once with us and automatically get integrations with hundreds of other apps (750+ and growing).

I love and use several Panic products (Transmit, Prompt, Firewatch) and hope this end-of-life enables them to spend more time on new ideas.

View on HN

I found Zapier's integrations to be very narrow (e.g., hitting only one major use case at best) and cost prohibitive.

The concept sounded really cool, but then when it came down to actually using in production I had a hard time arguing with myself: "Eh, why not spend a night or two just building out this integration ourselves the exact way we want instead of wasting that much money on something that doesn't even completely fit the bill."

The subscription SaaS cost quickly supersedes the developer cost in a matter of a few months, although I suppose it could have some niche uses with a proof of concept/MVP.

You have a valid reason for not using Zapier today because you know how to write code (though some still do).

The majority of Zapier users are not engineers. They can't make that trade-off decision. Their trade-off is instead "should I use Zapier or just not do X at all?"

I'd guess many of Panic board's users chose the latter (not do X at all) given they cited lack of integrations as one reason for shutting down.

Well, I know how to write code and I still barely use Zapier because I value my time, IFTTT is "good enough" for most things, and also porting all my IFTTT recipes would cost me -way- too much a month.

Not a complaint about Zapier, mind, just that "you can code" isn't a great argument on this point.

Our non-developers love it. They tend to set up quick trials of new integrations, often building new processes with Google Forms, Zapier and Slack. The ones that become important/expensive enough we build in house, but most never make it that far and would be a waste of engineering resources to build.

Are you seriously telling the founder of Zapier that the product sucks and will never see success? I think Zapier is amazing and I know plenty of people that feel the same. It's not some obscure thing.

No, I'm sharing why Zapier didn't work for me. Hope they have all the success in the world.

I don't think this response is warranted, and this is not what he's saying. He's merely illustrating why it didn't work for him.

A night or two. Thats how a months long project starts.

> The subscription SaaS cost quickly supersedes the developer cost in a matter of a few months

That is emphatically not my experience. Developers consistently underestimate how expensive they are and how long things take to do.

Even if you ignore the up-front development cost, I probably spend more than 15 minutes a month on maintaining any integrations I write. That time alone more than covers Zapier's cost ($20/m to start).

Reply on news.ycombinator.com